Abstract The biological basis of PD with dementia (PDD) is heterogeneous, involves multiple cognitive domains, and is challenging to characterize. Fronto-striatal executive/attentional impairments are core early features and are influenced by dopaminergic neurotransmission. PDD arises when a distinct visuospatial cognitive syndrome-- thought to be mediated by non-dopaminergic changes?begins to substantially affect the integrity of posterior cortical regions. The pathological substrates underlying these impairments are diverse and involve the toxic effects of misfolded neuronal alpha-synuclein along with non-synuclein pathologies such as Amyloid-beta (Abeta) plaques, tau aggregates, and vascular cognitive impairment. The heterogeneous nature of these pathologies and their clinical manifestations is summarized by the ?dual-hit? hypothesis: a well-accepted but never-before-tested principle in current models of PD?PDD pathogenesis. We propose a proof-of-concept PD trial aimed at delaying visuospatial cognitive decline, a critical component of PDD. In PD, low-range cortical Abeta plaque levels?well below the Alzheimer?s disease (AD) diagnostic threshold?associate with serotonin terminal losses. This specific serotonin-Abeta characteristic is true of PD but not AD. Our published multicenter PD observational findings show that selective serotonin reuptake inhibitors (SSRIs) associate with lower PDD conversion risk and altered CSF Abeta-42 levels. Our preliminary data, using Abeta Pittsburgh compound B (PiB) positron emission tomography (PET) in PD, show less 2-year visuospatial cortex Abeta plaque accrual and reduced visuospatial cognitive decline in subjects on serotoninergic medications, perhaps modifying a key arbiter of PD?PDD risk. Our long-term vision is a PD?PDD prevention trial, in which citalopram would be initiated prior to Abeta-linked cognitive decline. Our central hypothesis is that citalopram use in PD will reduce visuospatial cortex Abeta plaque accrual, leading to an amelioration of visuospatial cognitive decline linked to PDD. Our objective is to test this hypothesis in a proof-of-concept trial of citalopram 20mg daily over 2 years in PD subjects (age ?65) without depression (n=58). In Aim 1, we will test differences in 2-year PiB PET accrual within the visuospatial cortex between citalopram and placebo groups. In Aim 2, we will measure the impact of this visuospatial cortex Abeta intervention upon visuospatial cognitive function. In Aim 3, we will test whether citalopram relative to placebo confers a minimal clinically important difference of benefit as assessed by a common measure of multi-domain cognition, the Montreal Cognitive Assessment, linked to PDD diagnosis. At the conclusion of this study, we will have well-powered dataset to assess our rigorous Go/No-Go criteria for this safe and generalizable PDD prevention therapy. We will also conduct analyses (Aims 1, 2, & 3) that will help to either substantiate or invalidate the dual-hit model of PD?PDD pathogenesis, a significant contribution to the PD field with implications for other Alzheimer?s disease and related dementia (ADRD) conditions as well.